Narrative Nubs: Systematic Sketches of Story Systems
Proposal Type
Individual Talk
Location
Hypertexts & Fictions
Start Date
July 2026
End Date
July 2026
Abstract
Since the 1960s, programmers have developed story generators to learn about language and cognition, to explore the creative writing process and creativity itself, and for artistic and literary purposes. Regardless of motivation, many of these — among them TALE-SPIN, GESTER, MINSTREL, Grandmother, MEXICA, DAYDREAMER, and several innovative smaller-scale systems — are compelling as electronic literature. Readers have been invited to learn about these and other systems via their outputs (Montfort and Bertram 2024); another way to encounter them is by studying code, running it, and seeing these systems work. What code is available, however, is seldom complete. When it is, it may not run on contemporary computers. Even then, some programs may only be available to the original programmers.
Sketching existing software, creating miniatures that abstract the essential features of programs, and reconstructing or reimplementing programs has also been done for decades. The miniatures concept was developed at Yale (Schank and Riesbeck 1981); it has been theorized further recently (Sack 2023) by a researcher who created an important early TALE-SPIN reimplementation in 1992 and collaborated on a more recent one. Researchers have reimplemented story generators at UC Santa Cruz, Smith, and UNAM. This sort of software development is not a preservation or archiving effort; it is programming as inquiry and a type of rational reconstruction (Tearse et. al 2012) that offers insight into the detailed workings and even cosmology of a system.
We at MIT and the Center for Digital Narrative are undertaking an open-ended project to create systematic “narrative nubs” (Montfort et al. in preparation). As with miniatures and most reimplementations, these nubs (as in “the nub of the argument”) are abstractions that capture core ideas, not exact recreations. The first release of the project features four nubs, for Story Machine, TAILOR, GESTER, and “Through the Park,” representing very different approaches to story generation (Montfort 2026). We use the widespread language Python 3, share modules when appropriate, and code in a consistent style. We aim to facilitate additional research, teaching, and literary art; the process of nub development has already yielded insights into story generation.
References
Bertram, Lillian-Yvonne and Nick Montfort. Output: An Anthology of Computer-Generated Text, 1953–2023. MIT Press: 2024.
Montfort, Nick and contributors. Narrative Nubs. Version 1.0. May 21, 2026. https://codeberg.org/nickmontfort/narrative-nubs/releases
Montfort, Nick, Jakob Kusnick, Louis Escouflaire, Daniel Villagran, Michelle Mo and Aadya Sharma. “Narrative Nubs: Reimplementation as Inquiry into Story Generation’s History.” In preparation.
Sack, Warren. “Miniatures, Demos and Artworks: Three kinds of computer program, their uses and abuses.” Talk at MIT, October 13, 2023.
Schank, Roger C. and Christopher K. Riesbeck. Inside Computer Understanding: Five Programs Plus Miniatures. Psychology Press, 1981.
Brandon Tearse, Peter Mawhorter, Michael Mateas and Noah Wardrip-Fruin. “Lessons Learned from a Rational Reconstruction of MINSTREL.” Proceedings of the AAAI Conference on Artificial Intelligence 26: 1, 249–255. 2012.
Narrative Nubs: Systematic Sketches of Story Systems
Hypertexts & Fictions
Since the 1960s, programmers have developed story generators to learn about language and cognition, to explore the creative writing process and creativity itself, and for artistic and literary purposes. Regardless of motivation, many of these — among them TALE-SPIN, GESTER, MINSTREL, Grandmother, MEXICA, DAYDREAMER, and several innovative smaller-scale systems — are compelling as electronic literature. Readers have been invited to learn about these and other systems via their outputs (Montfort and Bertram 2024); another way to encounter them is by studying code, running it, and seeing these systems work. What code is available, however, is seldom complete. When it is, it may not run on contemporary computers. Even then, some programs may only be available to the original programmers.
Sketching existing software, creating miniatures that abstract the essential features of programs, and reconstructing or reimplementing programs has also been done for decades. The miniatures concept was developed at Yale (Schank and Riesbeck 1981); it has been theorized further recently (Sack 2023) by a researcher who created an important early TALE-SPIN reimplementation in 1992 and collaborated on a more recent one. Researchers have reimplemented story generators at UC Santa Cruz, Smith, and UNAM. This sort of software development is not a preservation or archiving effort; it is programming as inquiry and a type of rational reconstruction (Tearse et. al 2012) that offers insight into the detailed workings and even cosmology of a system.
We at MIT and the Center for Digital Narrative are undertaking an open-ended project to create systematic “narrative nubs” (Montfort et al. in preparation). As with miniatures and most reimplementations, these nubs (as in “the nub of the argument”) are abstractions that capture core ideas, not exact recreations. The first release of the project features four nubs, for Story Machine, TAILOR, GESTER, and “Through the Park,” representing very different approaches to story generation (Montfort 2026). We use the widespread language Python 3, share modules when appropriate, and code in a consistent style. We aim to facilitate additional research, teaching, and literary art; the process of nub development has already yielded insights into story generation.
References
Bertram, Lillian-Yvonne and Nick Montfort. Output: An Anthology of Computer-Generated Text, 1953–2023. MIT Press: 2024.
Montfort, Nick and contributors. Narrative Nubs. Version 1.0. May 21, 2026. https://codeberg.org/nickmontfort/narrative-nubs/releases
Montfort, Nick, Jakob Kusnick, Louis Escouflaire, Daniel Villagran, Michelle Mo and Aadya Sharma. “Narrative Nubs: Reimplementation as Inquiry into Story Generation’s History.” In preparation.
Sack, Warren. “Miniatures, Demos and Artworks: Three kinds of computer program, their uses and abuses.” Talk at MIT, October 13, 2023.
Schank, Roger C. and Christopher K. Riesbeck. Inside Computer Understanding: Five Programs Plus Miniatures. Psychology Press, 1981.
Brandon Tearse, Peter Mawhorter, Michael Mateas and Noah Wardrip-Fruin. “Lessons Learned from a Rational Reconstruction of MINSTREL.” Proceedings of the AAAI Conference on Artificial Intelligence 26: 1, 249–255. 2012.
Bio
Nick Montfort’s work includes ten computer-generated books from seven presses, the collaborations Start Me Up, The Deletionist, and Sea and Spar Between, and many other digital projects. His latest poetry book, All the Way for the Win, is composed entirely of three-letter words. His MIT Press books include The Future and two co-edited volumes, The New Media Reader and Output: An Anthology of Computer-Generated Text, 1953–2023. He’s a professor at MIT, principal investigator in the University of Bergen’s Center for Digital Narrative, and director of The Trope Tank. He lives in New York City.